Menu

Human-AI Interaction

We're serving over 108+ low resource languages through our innovative products.

Reteena

Alzheimer's patients lack tools to preserve their memories and track cognitive changes over time, highlighting the need for adaptive digital twins to enhance care and research while ensuring privacy and ethics. Reteena's proposed pipeline will use LLMs, VLMs, and Human-AI Interaction to capture and recall patients' memories unique personality, habits, and memories. This will help aid researchers in studying Alzheimer's progression and testing interventions.

Prompt Strategies for Summarizing Doctor-Patient Dialogues with LLMs

We're working with Chris Emezue, the founder of the startup Lanfrica, to explore innovative methods for summarizing doctor-patient encounter dialogues using large language models. This system aims to generate concise and actionable summaries of doctor-patient interactions, enhancing communication and record-keeping in medical systems.

EduLang

EduLang, a multilingual library app, bridges the language gap through storytelling. The personalized book library provides learners with culturally sensitive, relatable and learning level appropriate content. Our app uses interactive multilingual books to help learners in grades K-5 from non-English speaking families learn English. Edulang is targeted towards low-resource language speakers with an aim to enable English language learning while continuing to preserve the learner's native language proficiency.

Prompt Strategies for Summarizing Doctor-Patient Dialogues with LLMs

We're working with Chris Emezue, the founder of the startup Lanfrica, to explore innovative methods for summarizing doctor-patient encounter dialogues using large language models. This system aims to generate concise and actionable summaries of doctor-patient interactions, enhancing communication and record-keeping in medical systems.

Yekola

We're building the MVP for Yekola, an early-stage startup that aims to learn, teach, and preserve low-resource languages. Yekola currently has over 3000+ users on their waitlist and we plan to launch the MVP during our cohort.

PuzzLing

PuzzLing combines the full system for the CALL(Computer Assisted Language Learning) platform especially focusing on low-resource languages, which includes language scoring and feedback functions. With the support from the latest language processing toolkit of the Neural Space, we aim to give a general evaluation and retrieve the error places that the testers can improve.

EduLang

EduLang, a multilingual library app, bridges the language gap through storytelling. The personalized book library provides learners with culturally sensitive, relatable and learning level appropriate content. Our app uses interactive multilingual books to help learners in grades K-5 from non-English speaking families learn English. Edulang is targeted towards low-resource language speakers with an aim to enable English language learning while continuing to preserve the learner's native language proficiency.

Prompt Strategies for Summarizing Doctor-Patient Dialogues with LLMs

We're working with Chris Emezue, the founder of the startup Lanfrica, to explore innovative methods for summarizing doctor-patient encounter dialogues using large language models. This system aims to generate concise and actionable summaries of doctor-patient interactions, enhancing communication and record-keeping in medical systems.

Yekola

We're building the MVP for Yekola, an early-stage startup that aims to learn, teach, and preserve low-resource languages. Yekola currently has over 3000+ users on their waitlist and we plan to launch the MVP during our cohort.

PuzzLing

PuzzLing combines the full system for the CALL(Computer Assisted Language Learning) platform especially focusing on low-resource languages, which includes language scoring and feedback functions. With the support from the latest language processing toolkit of the Neural Space, we aim to give a general evaluation and retrieve the error places that the testers can improve.